ABSTRACT Four million neonatal deaths occur annually and account for 40% of under-five mortality. These deaths occur predominately in low-resource settings where the majority of infants are born at home. In settings of high neonatal mortality risk, up to 50% of deaths are attributable to infection. The long-term objective of our research program is to identify simple, affordable, and effective interventions that can be delivered at the community level in low-resource settings to reduce neonatal and early infant mortality. There is increasing evidence of the role of the neonatal skin and the potential for newborn massage with topical emollients in providing protection from invasive pathogens. This research project focuses on evaluating the impact of newborn massage with sunflower seed oil on neonatal mortality and infections, relative to massage with mustard oil, which is provided almost universally to tens of millions of newborns in South Asia each year. The primary specific aim (Aim 1) is to compare the neonatal mortality risk among infants randomized to receive repeated full-body massage during the neonatal period with either sunflower seed oil or mustard seed oil. Secondary aims include a between-group comparison of mortality risk among preterm infants (Aim 2) and a comparison of the incidence of probable severe disease (sepsis) between the sunflower and mustard oil groups (Aim 3). The study population will be all live-born infants delivered within our community-based research site in Sarlahi District of southern Nepal. This site is divided into 430 communities that will be randomized to either sunflower or mustard oil. Pregnant women within the catchment area will be recruited mid-pregnancy, and provided with a set of common basic antenatal interventions (tetanus toxoid, clean delivery kit, iron-folate supplements, and basic educational messages). Women will also be provided with either sunflower seed oil or mustard oil, depending on the random allocation of their cluster of residence. The oil, sufficient in quantity to provide full-body massage three times daily for 28 days, will be provided by locally resident, village-level workers along with guidelines on the use of the oil and actual conduct of the massage. All women will be followed to pregnancy outcome. After birth of the infant, the local project workers will visit each day for the first week of life to continue promoting the massage and measuring compliance. A team of data collectors will visit all infants at home on days 1, 3, 7, 10, 14, and 28 to record vital status and basic signs of morbidity. The primary outcome is mortality within 28 days of birth. For the secondary morbidity outcome, a clinical sign- based algorithm for defining probable sepsis will be utilized. To detect a 20% reduction in all-cause neonatal mortality with 90% power and Type I error of 5%, the required sample size per group is 13,601. This sample size accounts for the clustered design and 5% loss to follow up. Given the population and crude birth rate in the study setting, recruitment of the required sample is expected to take 3.4 years.